Register or Log In

International Society for Industrial Process Tomography

12th World Congress on Industrial Process Tomography

Visualization of Particle Distributions in Centrifuge by Electrical Resistance Tomography with Optimized Linear-Arrayed Electrode Sensor (LAES-ERT) based on Genetic Algorithm

Z. F. GAO1, T. ZHAO2*, A. ZHANG2 , Z. Q. LIU2

1School of Mechanical and Electrical Engineering, Lanzhou University of Technology, Lanzhou, China
2School of Mechanical and Presicion Instrumental Engineering, Xi’an University of Technology, Xi’an, China

*Email: tongzhao@xaut.edu.cn


ABSTRACT

This study proposed a novel visualization method to obtain the particle distributions inside the centrifuge based on electrical resistance tomography with linear-arrayed electrode sensor (LAES-ERT). An ERT sensor with linear arrayed electrodes along the axil direction injects the current and collects the voltage, based on which the conductivity distribution in the centrifuge is reconstructed, and the thickness of the particle distribution is detected. In order to improve the accuracy of reconstructed conductivity for accurate visualization of particle distribution, the parameters of ERT sensor are optimized based on genetic algorithm in this study. Firstly, a 3D field model with different electrode parameters, including electrode number (N), electrode length (L), and electrode width (W), is used to simulate the electrical field and reconstruct the conductivity. Secondly, a feedforward neural network is used to predict the nonlinear relationship between electrode parameters and conductivity evaluation parameters. Thirdly, the electrode parameters are optimized by combining the feedforward neural network and the genetic algorithm. The experiment results show that the linear-arrayed electrode sensor with optimum electrode parameters enables to detect the particle distribution successfully, the thicknesses of particle in 3 mm, 6 mm, and 9 mm are accurately detected as 3.3 mm, 5.2 mm, and 8.9 mm, respectively. In conclusion, the LAES-ERT is able to detect the thickness of particle distributions in the centrifuge, which helps to monitoring the separation process of centrifuge.


Keywords: Decanter Centrifuge, Electrical resistance tomography, Genetic algorithm, Linear-arrayed electrode sensor, Particle distribution detection

Industrial Application:

Sign-in to access the full text